Commit Graph

12 Commits

Author SHA1 Message Date
Chummy
b721754ead feat(codex): add websocket-first transport selection 2026-02-28 13:48:41 +08:00
argenis de la rosa
281236a94d feat(identity): add openclaw extra_files support 2026-02-27 15:15:39 -05:00
argenis de la rosa
e7e513d7ec fix(onboard): tailor memory scaffolding by backend 2026-02-26 21:48:35 -05:00
guitaripod
0a7931e73e
fix(agent): add channel media markers to system prompt (#1697)
The system prompt had no documentation of channel media markers
([Voice], [IMAGE:], [Document:]), causing the LLM to misinterpret
transcribed voice messages as unprocessable audio attachments instead
of responding to the transcribed text content.

Co-authored-by: Argenis <theonlyhennygod@gmail.com>
2026-02-24 19:07:27 -05:00
Chummy
e081010983 feat(skills): add configurable compact skills prompt injection 2026-02-21 00:00:51 +08:00
Chummy
3733856093 Fix skill instruction/tool injection in system prompts 2026-02-20 02:16:41 +08:00
Chummy
d9a94fc763 fix(skills): escape inlined skill XML content 2026-02-20 01:28:49 +08:00
Edvard
8a4da141d6 fix(skills): inject skill prompts and tools into agent system prompt
Skill prompts and tool definitions from SKILL.toml were parsed and stored
correctly but never included in the agent's system prompt. Both prompt-building
paths (channels/mod.rs and agent/prompt.rs) only emitted skill metadata (name,
description, location), telling the LLM to "read" the SKILL.toml on demand.
This caused the agent to attempt manual file reads that often failed, leaving
skills effectively ignored.

Now both paths inline <instructions> and <tools> blocks inside each <skill>
XML element, so the agent receives full skill context without extra tool calls.

Closes #877

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-02-20 01:28:49 +08:00
Chummy
d6dca4b890 fix(provider): align native tool system-flattening and add regressions 2026-02-19 17:44:07 +08:00
YubinghanBai
48eb1d1f30 fix(agent): inject full datetime into system prompt and allow date command
Three related agent UX issues found during MiniMax channel testing:

1. DateTimeSection injected only timezone, not the actual date/time.
   Models have no reliable way to know the current date from training
   data alone, causing wrong or hallucinated dates in responses.
   Fix: include full timestamp (YYYY-MM-DD HH:MM:SS TZ) in the prompt.

2. The `date` shell command was absent from the security policy
   allowed_commands default list. When a model tried to call
   shell("date") to get the current time, it received a policy
   rejection and told the user it was "blocked by security policy".
   Fix: add "date" to the default allowed_commands list. The command
   is read-only, side-effect-free, and carries no security risk.

3. (Context) The datetime prompt fix makes the date command fallback
   largely unnecessary, but the allowlist addition ensures the tool
   works correctly if models choose to call it anyway.

Non-goals:
- Not changing the autonomy model or risk classification
- Not adding new config keys

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-02-19 17:44:07 +08:00
Devin AI
44fa7f3d3d fix(agent): include workspace files when AIEOS identity is configured
Remove early return in IdentitySection::build() that caused AGENTS.md,
SOUL.md, and other workspace files to be silently skipped when AIEOS
identity loaded successfully. Both AIEOS identity and workspace files
now coexist in the system prompt.

Closes zeroclaw-labs/zeroclaw#856

Co-Authored-By: Kristofer Mondlane <kmondlane@gmail.com>
2026-02-19 15:24:58 +08:00
mai1015
b341fdb368 feat: add agent structure and improve tooling for provider 2026-02-17 01:01:56 +08:00